Multi-attribute Probability Caching Algorithm in Named Data Network

Abstract

Since entering the new century, Internet technology has developed rapidly and has become an important infrastructure in today’s society. In recent years, with the popularization of optical fiber applications, the network bandwidth has been greatly improved, and the growth rate of network traffic has also shown explosive growth. The traditional TCP/IP network has exposed many serious problems in the face of increasing traffic. In particular, the issues of scalability, reliability, security, flexibility and mobility are more prominent. In order to solve the shortcomings of the existing Internet architecture, the current domestic and international schemes mainly include "improvement" and "revolution". Improvement is the gradual improvement of the deficiencies of the existing Internet system. However, this method cannot essentially solve these deficiencies, and it is difficult to meet the needs of future network development. The revolution is to redesign the information-centric network as the Internet system structure, completely changing the existing Internet system in terms of scalability, security and mobility, in order to adapt to the needs of future social development. Named-Data Networking (NDN) in many information-centric network architectures is a network architecture with excellent research and development. This article focuses on the research of NDN network caching technology. In terms of caching decisions, a multi-attribute probabilistic caching algorithm is proposed, which can effectively reduce the cache of redundant data in routing nodes in NDN networks and improve the utilization of routing node cache space, reduce network delay, increase the cache hit rate of routing nodes, and improve the overall performance of the NDN network.